Hello, below is the assignment details. It is written in Python. I will also need to include the best performance on the test set in terms of F1 score.

Directions: For this assignment, you need to reimplement the famous AlexNet model. Below is a detailed instruction of what you may need to do. - Dataset Preparation - You need to load the dataset using torchvision.datasets.CIFAR100. - More information about the function can be found at: - Please notice that CIFAR has two versions, CIFAR10 and CIFAR100. You need to use CIFAR 100 for this assignment. Train/Test Split - With torchision.datasets.CIFAR100, you don't need to split the dataset by yourself. But, you do need to set train=True to load the training set, and set train=False to load the test set. - Model Development AlexNet - You need to reimplement the famous AlexNet model by yourself according to the architecture on the right. - Notice that the output layer of a neural network classification model must equal the number of classes. Since AlexNet was developed for the ImageNet challenge, the outsize was I Finta set to 1000 . However, for your case, you need to change the output shape according to the CIFAR 100 dataset (i.e., set the output shape equals the number of classes in CIFAR100). - Train/Test Dense: 4096 fully connected neurons Polu, dropout p=0.5 Dense: 1000 fully connected neurons Output: 1 of 1000 classes - For this project, you will only use two sets to develop your model. You will use the test set provided by PyTorch as the dev set. - (Optional) Hyperparameters are extremely important to neural network performance. Some examples of hyperparameters may include training epoch, learning rate, optimizers, schedulers, etc. To achieve the best performance, you may want to try different hyperparameters